metadata
license: mit
base_model: microsoft/layoutlm-base-uncased
tags:
- generated_from_trainer
model-index:
- name: layoutlm-custom_no_text
results: []
layoutlm-custom_no_text
This model is a fine-tuned version of microsoft/layoutlm-base-uncased on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.1523
- Noise: {'precision': 0.8811544991511036, 'recall': 0.8994800693240901, 'f1': 0.8902229845626072, 'number': 577}
- Signal: {'precision': 0.8675721561969439, 'recall': 0.8856152512998267, 'f1': 0.8765008576329331, 'number': 577}
- Overall Precision: 0.8744
- Overall Recall: 0.8925
- Overall F1: 0.8834
- Overall Accuracy: 0.9664
Model description
More information needed
Intended uses & limitations
More information needed
Training and evaluation data
More information needed
Training procedure
Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 3e-05
- train_batch_size: 8
- eval_batch_size: 8
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 15
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Noise | Signal | Overall Precision | Overall Recall | Overall F1 | Overall Accuracy |
|---|---|---|---|---|---|---|---|---|---|
| 0.3886 | 1.0 | 18 | 0.2452 | {'precision': 0.6213235294117647, 'recall': 0.58578856152513, 'f1': 0.6030330062444246, 'number': 577} | {'precision': 0.6323529411764706, 'recall': 0.5961871750433275, 'f1': 0.6137377341659233, 'number': 577} | 0.6268 | 0.5910 | 0.6084 | 0.8992 |
| 0.1673 | 2.0 | 36 | 0.1441 | {'precision': 0.7667269439421338, 'recall': 0.7348353552859619, 'f1': 0.7504424778761062, 'number': 577} | {'precision': 0.7450271247739603, 'recall': 0.7140381282495667, 'f1': 0.7292035398230089, 'number': 577} | 0.7559 | 0.7244 | 0.7398 | 0.9356 |
| 0.0959 | 3.0 | 54 | 0.1168 | {'precision': 0.8131487889273357, 'recall': 0.8145580589254766, 'f1': 0.8138528138528138, 'number': 577} | {'precision': 0.7941176470588235, 'recall': 0.7954939341421143, 'f1': 0.7948051948051947, 'number': 577} | 0.8036 | 0.8050 | 0.8043 | 0.9510 |
| 0.0622 | 4.0 | 72 | 0.1166 | {'precision': 0.8402061855670103, 'recall': 0.8474870017331022, 'f1': 0.8438308886971526, 'number': 577} | {'precision': 0.8333333333333334, 'recall': 0.8405545927209706, 'f1': 0.8369283865401207, 'number': 577} | 0.8368 | 0.8440 | 0.8404 | 0.9591 |
| 0.0424 | 5.0 | 90 | 0.1325 | {'precision': 0.8476027397260274, 'recall': 0.8578856152512998, 'f1': 0.8527131782945737, 'number': 577} | {'precision': 0.839041095890411, 'recall': 0.8492201039861352, 'f1': 0.8440999138673558, 'number': 577} | 0.8433 | 0.8536 | 0.8484 | 0.9586 |
| 0.031 | 6.0 | 108 | 0.1167 | {'precision': 0.8720136518771331, 'recall': 0.8856152512998267, 'f1': 0.878761822871883, 'number': 577} | {'precision': 0.8583617747440273, 'recall': 0.8717504332755632, 'f1': 0.8650042992261393, 'number': 577} | 0.8652 | 0.8787 | 0.8719 | 0.9628 |
| 0.0213 | 7.0 | 126 | 0.1339 | {'precision': 0.8610634648370498, 'recall': 0.8700173310225303, 'f1': 0.8655172413793105, 'number': 577} | {'precision': 0.855917667238422, 'recall': 0.8648180242634316, 'f1': 0.860344827586207, 'number': 577} | 0.8585 | 0.8674 | 0.8629 | 0.9608 |
| 0.0159 | 8.0 | 144 | 0.1335 | {'precision': 0.8692699490662139, 'recall': 0.8873483535528596, 'f1': 0.8782161234991425, 'number': 577} | {'precision': 0.8590831918505942, 'recall': 0.8769497400346621, 'f1': 0.8679245283018868, 'number': 577} | 0.8642 | 0.8821 | 0.8731 | 0.9630 |
| 0.0117 | 9.0 | 162 | 0.1489 | {'precision': 0.8686006825938567, 'recall': 0.8821490467937608, 'f1': 0.8753224419604471, 'number': 577} | {'precision': 0.8600682593856656, 'recall': 0.8734835355285961, 'f1': 0.8667239896818572, 'number': 577} | 0.8643 | 0.8778 | 0.8710 | 0.9622 |
| 0.011 | 10.0 | 180 | 0.1593 | {'precision': 0.8623063683304647, 'recall': 0.8682842287694974, 'f1': 0.8652849740932642, 'number': 577} | {'precision': 0.8519793459552496, 'recall': 0.8578856152512998, 'f1': 0.854922279792746, 'number': 577} | 0.8571 | 0.8631 | 0.8601 | 0.9600 |
| 0.0094 | 11.0 | 198 | 0.1336 | {'precision': 0.8896434634974533, 'recall': 0.9081455805892548, 'f1': 0.8987993138936535, 'number': 577} | {'precision': 0.8760611205432938, 'recall': 0.8942807625649913, 'f1': 0.8850771869639794, 'number': 577} | 0.8829 | 0.9012 | 0.8919 | 0.9686 |
| 0.0066 | 12.0 | 216 | 0.1357 | {'precision': 0.8928571428571429, 'recall': 0.9098786828422877, 'f1': 0.9012875536480687, 'number': 577} | {'precision': 0.8792517006802721, 'recall': 0.8960138648180243, 'f1': 0.8875536480686695, 'number': 577} | 0.8861 | 0.9029 | 0.8944 | 0.9692 |
| 0.0072 | 13.0 | 234 | 0.1528 | {'precision': 0.8830508474576271, 'recall': 0.902946273830156, 'f1': 0.8928877463581834, 'number': 577} | {'precision': 0.8711864406779661, 'recall': 0.8908145580589255, 'f1': 0.8808911739502999, 'number': 577} | 0.8771 | 0.8969 | 0.8869 | 0.9670 |
| 0.0061 | 14.0 | 252 | 0.1552 | {'precision': 0.8779661016949153, 'recall': 0.8977469670710572, 'f1': 0.8877463581833762, 'number': 577} | {'precision': 0.8661016949152542, 'recall': 0.8856152512998267, 'f1': 0.8757497857754927, 'number': 577} | 0.8720 | 0.8917 | 0.8817 | 0.9664 |
| 0.0054 | 15.0 | 270 | 0.1523 | {'precision': 0.8811544991511036, 'recall': 0.8994800693240901, 'f1': 0.8902229845626072, 'number': 577} | {'precision': 0.8675721561969439, 'recall': 0.8856152512998267, 'f1': 0.8765008576329331, 'number': 577} | 0.8744 | 0.8925 | 0.8834 | 0.9664 |
Framework versions
- Transformers 4.36.2
- Pytorch 2.1.0+cu121
- Datasets 2.16.1
- Tokenizers 0.15.0